1991-1992), the rate of subsequent incarceration episodes is 2 per hundred clients, a rate 78% lower than the subsequent incarceration rate (9 per hundred) of a matched group of clients with untreated alcohol and drug problems.

For those that successfully completed methadone treatment (discharged in 1991-1992), the rate of subsequent incarceration episodes is 2 per hundred clients, a rate 89% lower than the subsequent incarceration rate (18 per hundred) of a matched group of those with untreated alcohol and drug problems.

These lower incarceration rates result in substantially fewer days incarcerated per hundred clients as seen in the figure below.

FIGURE 5

Incarceration days per 100 clients in the three years subsequent to treatment19

By treatment modality

4000

Comparison Group

Treatment Group

For those that successfully completed outpatient treatment (discharged in 1991-1992), the rate of subsequent incarceration days is 927 per hundred clients, a rate 58% lower than the subsequent days of incarceration rate (2215 days per hundred clients) of a matched group of clients with untreated alcohol and drug problems.

For those that successfully completed residential treatment (discharged in 1991-1992), the rate of subsequent incarceration days is 360 days per hundred clients, a rate 75% lower than the subsequent days of incarceration rate (1434 days per hundred) of a matched group of clients with untreated alcohol and drug problems.

19 Effect of treatment completion vs. non-completion on subsequent arrests: F-10.6, p=.001.

For those that successfully completed methadone treatment (discharged in

1991-1992), the rate of subsequent incarceration days is 312 per hundred clients, a

rate 91% lower than the subsequent days of incarceration rate (3534 per hundred) of a matched group of those with untreated alcohol and drug problems.

EMPLOYMENT

Data on employment were gathered from AFS (Adult and Family Services) files,20 These data show that, for this sample, in the period subsequent to treatment, the wages paid to treatment completers were 65% higher than the wages paid to those who did not complete treatment. The advantage in subsequent wages is observed for clients in all service modules (outpatient, residential, and methadone) but is greatest in the methadone module where treatment completers eamed more than twice as much as non-completers.

20 Direct access to Employment Department records was not possible. Employment information was gathered through the AFS data system after clients were identified by a case number, name, and date of birth search. Thus, the employment data that we have reflect only those clients that could be tracked in the AFS files. (Six hundred ninety-seven cases, about two thirds of the sample, were trackable in AFS; 483 had employment eamings.) For all modules, trackable and non-trackable clients were fairly evenly distributed between treatment completers and non-completers. The individuals for whom information was found likely represent the poorest clients in the sample, and since our interest here is in assessing the expenditure of public assistance money on this population, these clients are clearly the most relevant to the study.

21 Effect of treatment completion vs. non-completion on subsequent earnings, controlling for prior earnings: F=17.0, p=.0001.

$100

$200

This increase in wages earned was due to two factors: an improvement in the earning power of clients (per week) and an improvement in the number of weeks worked in the period subsequent to treatment.

FIGURE 6

Improvement in earnings per week22

(Pre-treatment to post-treatment) By treatment modality

$300

Outpatient

Residential

Methadone

While treatment non-completers also had modest increases in their average eamings in the weeks in which they worked, treatment completers had far greater increases. This occurred across every module. For example, residential treatment noncompleters eamed about $49 more per week in the weeks they worked during the three year period subsequent to their incomplete treatment than they had in the two year period prior to treatment. However, residential treatment completers eamed about $178 more per week in the weeks they worked during the three year period subsequent to their treatment than they had in the two year period prior to treatment. This represents a 250% better eamings performance by treatment completers than by non-completers.

22 Effect of treatment completion vs. non-completion on improvement in eamings per week: F=3.9, p=.05.

Comparison G

Treatment Gro

Among the non-completers, those who had even some treatment had greater improvements in their average eamings in the subsequent period than those noncompleters whose only exposure to treatment was an intake session.